Paper Details

Name: Hardik Khandelwal, Akanksha Chauhan, Sachin Maurya, Nilesh Sharma, Neeraj Singh, Sahil Bhardwa
Country: India

Year: 2026
Volume: Volume No: 13, January, Year: 2026 (Special Issue)
Page Number: 922-931
DOI: https://doi.org/10.5281/zenodo.20623693

In this research we increase the precision of the diagnosis of mango leaf disease, we employed sophisticated machine learning techniques in this project. We created a model using the "Orange" program to categorise mango leaf disease into eight groups: healthy, sooty mould, powdery mildew, gall midge, die back, cutting weevil, bacterial canker, and anthracnose. The main goal of this study was to use analytics and historical data to improve the accuracy and speed of disease identification. We used logistic regression and neural networks to evaluate different machine learning approaches. Logistic regression gave an accuracy of over 99.2%. But the neural network gave accuracy of 99.3%. The overall framework performed well on all the evaluation metrics: F1 score, recall, Matthews correlation coefficient (MCC). The results of the neural network demonstrate its ability to identify intricate patterns in the data, which is crucial for the diagnosis of illness. This study demonstrates how machine learning can increase mango leaf disease identification accuracy and dependability. Our data driven approach using the ―Orange‖ software can handle large datasets efficiently and make crop health monitoring more accurate and reliable. Among the methods tested the neural network was the best and most precise method, it gives significant improvement in crop health supervision and disease management.